An observational study in Breast Cancer Early Stage Breast Cancer (Stage 1-3), sponsored by Casa Sollievo della Sofferenza IRCCS. Enrolling by invitation at 1 site in Italy. Open to female participants aged 18 Years and older. Per ClinicalTrials.gov, last updated 2026-01-30.
Sponsored by Casa Sollievo della Sofferenza IRCCS · Observational
This is a multicenter, observational validation study designed to evaluate the prognostic performance of the PORTENT algorithm in patients with early-stage breast cancer. The model integrates clinicopathological variables and the expression levels of two small non-coding RNAs (miR-3916 and miR-3613-5p) to estimate individual risk of developing distant metastases.
The primary objective is to assess the discriminatory ability of the PORTENT algorithm for predicting distant metastasis at predefined time points after diagnosis.
This multicenter retrospective observational study aims to clinically validate the PORTENT prognostic algorithm for predicting the risk of distant metastases in women with early-stage breast cancer. Female patients with Stage I-III disease from three independent cohorts with available residual tumor tissue and follow-up data will be included.
The primary endpoint of the study is the discriminatory performance of the PORTENT algorithm, assessed by the area under the receiver operating characteristic curve (AUC) for the prediction of distant metastases at 5 and 10 years from diagnosis.
The algorithm integrates established clinicopathological prognostic factors (tumor stage, histological grade, and Ki67-MIB1) with the expression levels of miR-3916 and miR-3613-5p. MicroRNA expression and target protein expression will be evaluated using RT-qPCR and immunohistochemistry.
Secondary and exploratory analyses will include model calibration assessed using calibration curves and the Integrated Calibration Index (ICI), as well as survival analyses (overall survival, progression-free survival, and metastasis-free survival) performed using Cox proportional hazards models.
The study includes three cohorts: CSS-cohort_2: 350 cases with ≥5 years follow-up, prospectively enrolled 2014-2019 within the TRANSCAN-BREMIR Study (Refs: Prot 116/CE 30/09/2014; Prot 140/CE 28/10/2014; Prot 150/CE 24/10/2018). External cohort: ≥450 patients with available residual tumor tissue, retrospectively selected; each center enrolls ≥75 patients (60 disease-free ≥5 years, 15 with metastatic progression ≥3 years). INT_Milan cohort: 300 Luminal A breast cancer cases treated with neoadjuvant therapy, with pre-treatment and/or surgical residual tumor tissue available.
Exclusion Criteria:
350 stage 1 to 3A patients will be selected from the BREMIR study (ID NCT06555354) prospective cohort.
Diagnostic Test: Validation of Prognostic tool
The External\_cohort will include patients enrolled retrospectively from the centers participating in the study. A total of at least 450 patients will be selected, with residual tumor tissue available at the respective Pathology Departments. Each center will enroll a minimum of 75 patients, comprising 60 who remain disease-free after at least 5 years of follow-up and 15 who experience metastatic progression after at least 3 years of follow-up.
Diagnostic Test: Validation of Prognostic tool
The INT\_Milan cohort includes 300 Luminal A breast cancer cases treated with neoadjuvant therapy, with available pre-treatment and/or surgical residual tumor tissue.
Diagnostic Test: Validation of Prognostic tool
Collection of Clinical History: Clinical data, including medical history, clinicopathological features (e.g., age, histology, receptor and nodal status), and follow-up information (presence or absence of metastasis, patient vital status), will be collected and entered into a dedicated platform by. Follow-up updates are scheduled by Month 48 of the project (December 31, 2025) to ensure timely and accurate patient outcome data. Laboratory Analysis: Residual tumor sections prepared by the Pathology Unit of each participating center will be sent to the Oncology Laboratory at CSS-IRCCS, where RNA will be extracted using standardized experimental procedures. Expression levels of miR-3916, miR-3613-5p, and their target genes will be analyzed by quantitative real-time PCR (RT-qPCR). Protein expression of target genes will be assessed via immunohistochemistry. Additionally, the extracted RNA will be analyzed at the Gerobiomics and Exposomics Laboratory of IRST IRCCS.
Discriminatory Performance of the Prognostic Algorithm (AUC) for Prediction of Distant Metastasis Within 5 Years From Diagnosis
Area under the Receiver Operating Characteristic (ROC) curve (AUC) with 95% confidence intervals for patient-level risk probabilities generated by the prognostic algorithm to predict the occurrence of distant metastases within 5 years after breast cancer diagnosis. Discrimination will be assessed using ROC curve analysis and DeLong's test.
Time frame: 5 years from diagnosis; interim analysis at March 2027.
Discriminatory Performance of the Prognostic Algorithm (AUC) for Prediction of Distant Metastasis Within 10 Years From Diagnosis
Area under the Receiver Operating Characteristic (ROC) curve (AUC) with 95% confidence intervals for patient-level risk probabilities generated by the prognostic algorithm to predict the occurrence of distant metastases within 10 years after breast cancer diagnosis. Discrimination will be assessed using ROC curve analysis and DeLong's test.
Time frame: 10 years from diagnosis; final analysis after completion of 10-year follow-up for all participants (expected by 2029).
Calibration of the Prognostic Model at 5 Years (Integrated Calibration Index)
Calibration of the prognostic algorithm will be evaluated at 5 years using the Integrated Calibration Index (ICI) and calibration plots to assess agreement between predicted and observed distant metastasis risk.
Time frame: 5 years from diagnosis; interim analysis at March 2027.
Calibration of the Prognostic Model at 10 Years (Integrated Calibration Index)
Calibration of the prognostic algorithm will be evaluated at 10 years using the Integrated Calibration Index (ICI) and calibration plots to assess agreement between predicted and observed distant metastasis risk.
Time frame: 10 years from diagnosis; final analysis after completion of 10-year follow-up (expected by 2029).
Difference in Discriminatory Performance Between Prognostic Models (ΔAUC) at 5 Years
Difference in AUC between the standard clinical prognostic model (clinicopathological variables only) and the extended model including miR-3916 and miR-3613-5p expression for prediction of distant metastases at 5 years. Statistical comparison will be performed using DeLong's test.
Time frame: 5 years from diagnosis; interim analysis at March 2027.
Difference in Discriminatory Performance Between Prognostic Models (ΔAUC) at 10 Years
Difference in AUC between the standard clinical prognostic model (clinicopathological variables only) and the extended model including miR-3916 and miR-3613-5p expression for prediction of distant metastases at 10 years. Statistical comparison will be performed using DeLong's test.
Time frame: 10 years from diagnosis; final analysis after completion of 10-year follow-up (expected by 2029).
Classification Performance of the Prognostic Algorithm at 10 Years (Sensitivity, Specificity, PPV, NPV)
Estimation of sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV), with 95% confidence intervals, for predefined probability thresholds corresponding to low (\<10%), intermediate (10-30%), and high (\>30%) 10-year distant metastasis risk categories.
Time frame: 10 years from diagnosis; final analysis after completion of 10-year follow-up (expected by 2029).
Association Between miR-3916 and miR-3613-5p Expression and Overall Survival
Association between miR-3916 and miR-3613-5p expression levels and overall survival, assessed using hazard ratios and 95% confidence intervals from Cox proportional hazards models.
Time frame: Up to 10 years from diagnosis; final analysis after completion of follow-up (expected by 2029).
Prognostic Algorithm Performance Within PAM50 Molecular Subgroups
Discriminatory performance and calibration of the prognostic algorithm within PAM50 molecular subgroups (Luminal A, Luminal B, HER2-enriched, Basal-like), evaluated using AUC and calibration metrics.
Time frame: 5 and 10 years from diagnosis; final analysis after completion of follow-up (expected by 2029).
Analytical Validity of miRNA Expression Assessment
Assessment of analytical validity of miR-3916 and miR-3613-5p expression measurement, including RNA yield, RT-qPCR amplification success rate, and assay reproducibility. This outcome assesses technical performance only.
Time frame: Data collection and analysis completed by March 2027.
Identification and Analytical Validation of miRNA Target Genes
Identification and analytical validation of molecular target genes regulated by miR-3916 and/or miR-3613-5p using in silico prediction, experimental confirmation, assay development, and RT-qPCR analytical validation.
Time frame: Data collection and analyses completed by March 2027.
Correlation of miRNA Target Gene Expression With Clinical Outcomes and Clinicopathological Features
Correlation of mRNA and protein expression levels of validated miRNA target genes with clinical outcomes (overall survival, progression-free survival, metastasis-free survival) and clinicopathological characteristics.
Time frame: Up to 10 years from diagnosis; final analysis after completion of follow-up (expected by 2029).
Improvement in Prognostic Risk Classification (NRI and IDI)
Net Reclassification Improvement (NRI) and Integrated Discrimination Improvement (IDI) assessing improvement in patient risk stratification achieved by extending the prognostic algorithm with miRNA target gene expression data.
Time frame: 5 and 10 years from diagnosis; final analysis after completion of follow-up (expected by 2029).
Documents are hosted by the registry — open the source record to download them.
Plan to share: Yes — Individual participant data (IPD), including raw tumor expression data, clinicopathological features, and follow-up information collected during this study, may be shared with participating centers and other qualified researchers upon formal request. Data sharing will be conducted under strict data use agreements to ensure confidentiality, data security, and compliance with all applicable ethical and legal standards. Access to IPD will be granted solely for scientifically valid, collaborative research purposes following approval by the study steering committee or data access committee. Requests must include a detailed research proposal, data management plan, and evidence of ethical approval. This policy promotes transparency, facilitates scientific collaboration, and aims to maximize the utility of collected data while protecting participant privacy and rights.
Supporting information: Study protocol, Sap, Analytic code
Eligibility is decided by the study team. Share this record with your doctor or contact the team directly.
No contact was published for this record. The registry link below has the sponsor’s details.
Get an email when the registry record changes — status, dates, results — or when someone posts here.
Sign in to followQuestions and observations about this study, from anyone following it. Not medical advice, and not a channel to the study team — their contact details are on the registry record.
Sign in to join the discussion. Reading takes no account; posting does. You choose a display name, and a pseudonym is the default.
Nothing here yet. If you are running this trial, taking part in it, or weighing whether to, this is the place to say so.
Casa Sollievo della Sofferenza IRCCS